🚀 go-pugleaf

RetroBBS NetNews Server

Inspired by RockSolid Light RIP Retro Guy

Article View: comp.periphs
Article #1489

Re: Why optical disks are slow to seek;

#1489
From: Steve Pope
Date: Mon, 17 Nov 1986 00:26
34 lines
2389 bytes
Because of a few statements made in the popular media, optical disks
have acquired the reputation of being "slow to seek".  While
this is basically true of a few currently available models, don't
expect it to be a permanent, universal problem.  Magnetic disks were
pretty slow to seek at the corresponding stage in their development.

One reason they don't spin the disk faster is because this would 
increase the energy needed to write bits on the disk (you'd have less
time to burn a bit).  This would increase the cost of the laser
-- under current economics.  Another reason is, the data rate is
already about the same as magnetic disks, and thus about the same as
what current interfaces can handle.  Most optical disks 
have a raw bit-rate between 2 and 25 MHz.

The big selling point of optical disks is their capacity, not their
access time.  Often, there is a single spiral track; others use 
concentric tracks.  Some spiral drives, when they seek, only seek to 
the approximate vicinity of the desired track, and then spin 
perhaps several full revolutions before reaching the desired sector.
This makes the average seek time a lot higher.  However, there's
no reason why in the future, optical drives will not be able to 
seek directly to the desired track.  The technology exists.

Media defects and resultant sector-redirection schemes also impact
seek time and throughput.

As a rule, the throughput of an optical disk is much better on long
multi-sector accesses than on a series of single-sector accesses
for a variety of reasons.  It's up to operating systems and applications
designers to realize this in order to use the current generation of 
drives efficiently.

steve pope (...ucbvax!spp)
Technical Details
Message-ID: <16395@ucbvax.BERKELEY.EDU>
Path: arc1-ca.pugleaf.net!utzoo!mnetor!seismo!rutgers!sri-unix!hplabs!ucbvax!oz.berkeley.edu!spp
References: <1128@tekig5.UUCP> <5100141@ccvaxa> <amdcad.13214>

Show Full Headers
Complete Article Headers
Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: arc1-ca.pugleaf.net!utzoo!mnetor!seismo!rutgers!sri-unix!hplabs!ucbvax!oz.berkeley.edu!spp
From: spp@oz.berkeley.edu (Steve Pope)
Newsgroups: comp.periphs
Subject: Re: Why optical disks are slow to seek;
Message-ID: <16395@ucbvax.BERKELEY.EDU>
Date: Sun, 16-Nov-86 19:26:35 EST
Article-I.D.: ucbvax.16395
Posted: Sun Nov 16 19:26:35 1986
Date-Received: Mon, 17-Nov-86 02:02:43 EST
References: <1128@tekig5.UUCP> <5100141@ccvaxa> <amdcad.13214>
Sender: usenet@ucbvax.BERKELEY.EDU
Reply-To: spp@oz.berkeley.edu.UUCP (Steve Pope)
Organization: University of California, Berkeley
Lines: 34